Jpn. J. Infect. Dis., 60, 397-399, 2007
Adult Pseudomonas aeruginosa Meningitis: High Incidence of Underlying
Medical and/or Postneurosurgical Conditions and High Mortality Rate
C. R. Huang, C. H. Lu, Y. C. Chuang, N. W. Tsai, C. C. Chang, S. F. Chen,
H. C. Wang1, C. C. Chien2 and W. N. Chang*
Department of Neurology, 1Department of Neurosurgery and 2Department of Diagnostic Pathology,
Chang Gung Memorial Hospital-Kaohsiung Medical Center, Chang Gung University College of
Medicine, Kaohsiung, Taiwan
(Received June 25, 2007. Accepted August 15, 2007)
SUMMARY: We analyzed the clinical and laboratory characteristics, therapeutic outcome and prognostic factors
of 25 cases of cerebrospinal fluid (CSF) culture-proven Pseudomonas aeruginosa adult bacterial meningitis
(ABM). Twelve P. aeruginosa strains, isolated from clinical CSF specimens, were tested for antibiotic suscep-
tibility. The 25 cases included 17 men and 8 women, aged 17 to 86 years (median = 51). Of the 25 cases of P.
aeruginosa ABM, 18 were the result of postneurosurgical infection and the other 7 were spontaneous infections.
The latter 7 cases had serious underlying medical conditions. The antibiotic susceptibility rates of the 12 strains
were as follows: ceftriaxone 16.7% (2/12), ceftazidime 91.7% (11/12), cefepime 83.3% (10/12), imipenem 83.3%
(10/12), meropenem 83.3% (10/12) and ciprofloxacin 66.7% (8/12). The therapeutic results showed an overall
mortality rate of 40% (10/25). The emergence of third-generation cephalosporin-resistant P. aeruginosa strains
cultured from clinical CSF specimens in recent years has resulted in a therapeutic challenge in the treatment of
In both eastern and western countries, Pseudomonas ceftriaxone, ceftazidime, cefepime, imipenem, meropenem
aeruginosa is not a common pathogen of bacterial meningitis and ciprofloxacin.
(1,2). Although large-scale studies of overall adult bacterial For comparative analysis, the 25 P. aeruginosa ABM cases
meningitis (ABM) have been conducted (1,2), P. aeruginosa were divided into two groups (fatal and non-fatal groups).
as a sole cause for ABM has been rarely examined in the Data including gender, type of acquisition of infection, type
literature. In this study, we analyzed the clinical and labo- of infection, underlying conditions, clinical manifestations
ratory features and the therapeutic outcomes of 25 mono- and therapeutic outcomes between these two patient groups
microbial P. aeruginosa ABM cases. We also examined the were analyzed by means of a chi-square test or Fisher’s exact
antimicrobial susceptibility of 12 P. aeruginosa strains test. The age difference between the two patient groups was
isolated from clinical cerebrospinal fluid (CSF) specimens. analyzed by means of Student’s t test. The consciousness lev-
During the study period (21 years, 1986 - 2006), 392 adult els of the two groups at the time of admission were analyzed
patients ( 17 years old) were determined to have culture- by means of the Mann-Whitney test. CSF data for the leuko-
proven bacterial meningitis. Of these 392 cases, 350 had cyte counts, glucose, total proteins and lactate concentrations
monomicrobial infections while the other 42 had poly- were logarithmically transformed to improved normality, and
microbial infections. Of the 350 cases with monomicrobial comparisons were made using Student’s t test. All analysis
infections, 29 cases had Pseudomonas ABM, and 25 of these was conducted using SAS (SAS Statistical Institute, Cary,
29 cases were cases of P. aeruginosa infection. The others N.C., USA).
were Pseudomonas mendocina (n = 2) and Pseudomonas The clinical and laboratory data of the 25 P. aeruginosa
stutzeri (n = 1), while one was not subtyped. Of the 42 cases ABM cases are listed in Tables 1 and 2. The underlying con-
of mixed infections, 12 involved Pseudomonas infection, and ditions of the seven cases with spontaneous infection were
all 12 of these were P. aeruginosa. The criteria for a definite systemic lupus erythematosus (SLE) (n = 1), otitis medium
diagnosis of P. aeruginosa meningitis included the follow- (n = 1), lung cancer with liver metastasis (n = 1), alcoholism
ing: (i) a positive CSF culture of P. aeruginosa, (ii) clinical with liver cirrhosis (n = 1), P. aeruginosa pneumonia with
features of meningitis and (iii) purulent CSF features. The respiratory failure (n = 1), diabetes mellitus (DM) with SLE
antibiotic susceptibility of the 12 isolated P. aeruginosa with end-stage renal disease (ESRD) (n = 1) and DM with
strains was tested using the Kirby-Bauer diffusion method ESRD (n = 1). In the 18 cases with a preceding neurosurgical
(BBL, Muller-Hinton II agars; Becton Dickinson Micro- state, the median time lag between the last neurosurgical pro-
biology Systems, Cockeysville, Md., USA) as described in cedure and the diagnosis of P. aeruginosa ABM was 20 days
the Clinical and Laboratory Standards (CLSI) guidelines for (3 - 720 days). The CSF data, as shown in Table 1, revealed
MICs. The antibiotics used in the susceptibility test included the features of purulent inflammation. The results of the anti-
microbial susceptibility tests are listed in Table 3. Overall,
*Corresponding author: Mailing address: Department of Neurol- 10 of the patients died and 15 survived, and only the differ-
ogy, Chang Gung Memorial Hospital-Kaohsiung, 123 Ta Pei ence in the level of the CSF leukocyte counts between the
Road, Niao Sung Hsiang, Kaohsiung Hsien 833, Taiwan. fatal and non-fatal groups was of statistical significance.
Tel: +886-7-7317123 ext. 2283, Fax: +886-7-7333816, E-mail: Six of the 12 cases whose CSF-isolated P. aeruginosa strains
firstname.lastname@example.org were used in the antibiotic susceptibility test died. The major
Table 1. Underlying medical conditions of the 25 adult Pseudomonas aeruginosa meningitis cases
OR 95% CI P-value
n = 15 n = 10
Age at meningitis (y) 39.9±18.3 53.3±20.9 0.328
Female 5 3 0.857 0.152 - 4.819 1
Male 10 7
Fever 11 8 0.917 0.123 - 6.825 1
Disturbed consciousness 1) 11 8 0.688 0.1 - 4.719 1
Seizure 7 4 1.313 0.259 - 6.643 1
Septic shock 0 3) 1 0.375 0.224 - 0.629 0.4
Hydrocephalus 7 4 1.313 0.259 - 6.643 1
CSF leakage 0 3) 1 0.375 0.224 - 0.629 0.4
Acquisition of infection
Community acquired 5 3 1.167 0.208 - 6.559 1
Nosocomial acquired 10 7
Types of infection
Spontaneous meningitis 4 3 0.848 0.144 - 4.99 1
Postneurosurgical meningitis 11 7
Underlying medical disorders
Diabetes mellitus 5 3 1.167 0.208 - 6.559 1
Alcoholism 1 1 0.643 0.036 - 11.631 1
End-stage renal disease 1 1 0.643 0.036 - 11.631 1
Systemic lupus erythematosus 1 2 0.286 0.022 - 3.669 0.543
Malignancy 1 1 0.643 0.036 - 11.631 1
Peripheral blood study
Bacteremia 3 5
Leukocytosis 2) 11 8 0.688 0.1 - 4.719 1
Glucose (mmol/L) 2.53 (0.06 - 8.21) 0.60 (0.22 - 6.50) 0.403
Total protein (g/L) 1.61 (0.19 - 27.59) 3.80 (1.18 - 6.45) 0.295
Lactate (mmol/L) 6.42 (0.22 - 21.70) 12.65 (2.79 - 23.43) 0.158
Leukocyte count ( 106/L) 0.14 (0.003 - 15.8) 0.85 (0.29 - 3.78) 0.024
: Glasgow coma scale <15.
: Leukocyte count > 10 109/L.
: Using a correction of 0.5 in every cell that contains a zero.
Table 2. Preceding neurosurgical events of the 18 adult Pseudomonas Table 3. Results of antibiotic susceptibility tests of 12 Pseudomonas
aeruginosa meningitis cases aeruginosa strains
Traumatic skull fracture and subarachnoid hemorrhange (SAH) (1) Susceptible test MICs (mg/dL)
Traumatic skull fracture (1) S I R Breakpoint MIC50 MIC90
Traumatic subdural hemorrhage (SDH) + SAH
Ceftriaxone 2 2 8 ≤8 64 128
+ ventriculoperitoneal (VP) shunt (1)
Ceftazidime 11 0 1 ≤8 4 8
Traumatic intracerebral hemorrhage (ICH) + SAH + SDH
+ craniectomy (1) Cefepime 10 1 1 ≤8 16 8
Traumatic epidural hemorrhage (EDH) + SDH + craniotomy (1) Imipenem 10 0 2 ≤4 1.5 8
Traumatic SDH + SAH + external ventricular drainage (EVD) (1) Meropenem 10 0 2 ≤4 0.5 8
Traumatic EDH + SDH + craniotomy (1) Ciprofloxacin 8 0 4 ≤1 1 16
s/p craniotomy (1) MICs, minimum inhibitory concentration; S, susceptible; I, intermediate;
Spontaneous ICH + EVD (1) R, resistant.
Arteriovenous malformation + ICH + VP shunt (1)
Aneurysm rupture + SAH + craniotomy +VP shunt (1)
(12/42) of ABM with polymicrobial infection. Among the
Brain tumor + craniotomy ± VP shunt (2)
implicated Pseudomonas spp. of monomicrobial ABM, P.
Chiari malformation + craniectomy (1)
aeruginosa was the most common, accounting for 86.2% (25/
Nasopharyngeal carcinoma with skull base invasion (2)
29) of the cases. Of the 25 P. aeruginosa ABM cases, 72% (18/
s/p VP shunt (2)
25) of the cases had a postneurosurgical state as the preced-
brain tumor, craniopharyngioma, astrocytoma; s/p, post state. ing event. In Taiwan, except for Klebsiella pneumoniae ABM,
which is commonly seen in DM patients as a community-
antibiotics used for the treatment of these 12 cases were acquired, spontaneous infection (2,3), the high incidence of
ceftazidime for the six surviving cases, and ceftazidime (n = a postneurosurgical state as the preceding event is also com-
4), cefepime (n = 1) and meropenem (n = 1) for the other six mon in other Gram-negative forms of ABM (2,4). Therefore,
expired cases. a positive culture result from the clinical specimens is the
This study revealed that Pseudomonas spp. accounted only practical method for further confirmation of P. aeruginosa
for 8.3% (29/350) of monomicrobial ABM and for 28.6% ABM.
The high incidence of third-generation cephalosporin- susceptibility and appropriate clinical microbiological advice.
resistant Gram-negative pathogens isolated from CSF speci- However, most of the P. aeruginosa ABM cases had complex
mens of ABM was previously reported in our study in 1999 preceding postneurosurgical conditions which may interfere
(5). In that study, Acinetobacter baumannii was found to with the analysis of the morbidity and mortality of ABM.
be the most notorious pathogen, and the others were K.
pneumoniae, Citrobacter freundii and Morganella morganii. REFERENCES
In the present study, the susceptibility rates of the tested P.
1. Durand, M.L., Calderwood, S.B., Weber, D.J., et al. (1993): Acute
aeruginosa strains to ceftriaxone, ceftazidime, cefepime, bacterial meningitis in adults. A review of 493 episodes. N. Engl. J.
imipenem, meropenem and ciprofloxacin were 16.7% (2/12), Med., 328, 21-28.
91.7% (11/12), 83.3% (10/12), 83.3% (10/12), 83.3% (10/ 2. Lu, C.H., Chang, W.N. and Chang, H.W. (2000): Adult bacterial
12) and 66.7% (8/12), respectively. These data revealed that meningitis in southern Taiwan: epidemiologic trends and prognostic
factors. J. Neurol. Sci., 182, 36-44.
third-generation cephalosporin-resistant P. aeruginosa strains 3. Lee, P.Y., Chang, W.N., Lu, C.H., et al. (2003): Clinical features and in
cultured from the clinical CSF specimens of ABM have vitro antimicrobial susceptibilities of community-acquired Klebsiella
emerged in recent years and were consistent with those pneumoniae meningitis in Taiwan. J. Antimicrob. Chemother., 51, 957-
reported by Lauderdale et al. (6) and Jones et al. (7). Although 962.
4. Lu, C.H., Chang, W.N., Chung, Y.C., et al. (1998): The prognostic
a judicious use of empiric antibiotic therapy is imperative in factor of adult Gram-negative bacillary meningitis. J. Hosp. Infect., 40,
the treatment of serious infectious diseases in order to avoid 27-34.
the development of resistance to antimicrobial agents, the 5. Lu, C.H., Chang, W.N., Chuang, Y.C., et al. (1999): Resistance to third
emergence of antibiotic-resistant, Gram-negative strains in generation cephalosporins resistance in adult Gram-negative bacillary
meningitis. Infection, 27, 208-211.
clinical settings were significantly correlated with the in- 6. Lauderdale, T.L., McDonald, L.C. and Shiau, Y.R. (2000): The status
creased consumption of antimicrobial agents (8,9). The emer- of antimicrobial resistance in Taiwan among gram-negative pathogens:
gence of clinical isolates of extended-spectrum cephalosporin- the Taiwan surveillance of antimicrobial resistance (TSAR) program,
resistant Gram-negative bacteria, including P. aeruginosa 2000. Diag. Microb. Infect. Dis., 48, 211-219.
strains, has resulted in a therapeutic challenge in the manage- 7. Jones, M.E., Draghi, D.C. and Karlowsky, J.A. (2004): Prevalence of
antimicrobial resistance in bacteria isolated from central nervous
ment of serious infectious diseases including nosocomial system specimens as reported by U.S. hospital laboratories from 2000
ABM (9,10). Although guidelines for the management of to 2002. Ann. Clin. Microbiol. Antimicrob., 3, 3.
ABM have been reported (11,12), the presence of extended- 8. Tenover, F.C. (2001): Development and spread of bacterial resistance
spectrum cephalosporin-resistant P. aeruginosa strains should to antimicrobial agents: an overview. Clin. Infect. Dis., 33 (Suppl. 3),
alert the clinicians who are dealing with the management of 9. Hsueh, P.R., Chen, W.H. and Luh, K.T. (2005): Relationships between
ABM. A more careful use of empiric antibiotics should antimicrobial use and antimicrobial resistance in Gram-negative bacteria
be considered, especially in the cases of nosocomial ABM causing nosocomial infections from 1991-2003 at a university hospital
patients in a postneurosurgical state. in Taiwan. Int. J. Antimicrob. Agents, 26, 463-472.
10. Ziai, W.C. and Lewin, J.J. (2006): Advances in the management of
This study revealed a high mortality rate (40%, 10/25) for central nervous system infections in the ICU. Crit. Care Clin., 22, 661-
P. aeruginosa ABM. Among the clinical and laboratory 694.
factors, only the level of the CSF leukocyte count had prog- 11. Tunkel, A.R., Hartman, B.J. and Kaplan, S.L. (2004): Practice guide-
nostic significance. Actually, there are many factors that may lines for the management of bacterial meningitis. Clin. Infect. Dis., 39,
influence the outcome of Gram-negative ABM (4), including 12. Cullen, M.M. (2005): An audit of the investigation and initial manage-
early diagnosis based on a high level of clinical attentiveness, ment of adults presenting with possible bacterial meningitis. J. Infect.,
compliance with updated and evidence-based guidelines, and 50, 120-124.
early and adequate antibiotic treatment based on the expected